oecd reviews of health care quality: turkey · pdf fileoecd reviews of health care quality...

Consult this publication on line at http://dx.doi.org/10.1787/9789264202054-en.

This work is published on the OECD iLibrary, which gathers all OECD books, periodicals and statistical databases. Visit www.oecd-ilibrary.org for more information.

isbn 978-92-64-20204-7 81 2013 20 1 P

OECD Reviews of Health Care Quality

TuRkEyRaising sTanDaRDs

At a time when ever more information is available about the quality of health care, the challenge for policy makers is to better understand the policies and approaches that sit behind the numbers. Turkey is the latest country report in this OECD series evaluating the quality of health care across countries – whether care is safe, effective and responsive to patients’ needs. OECD Reviews of Health Care Quality examine what works and what does not work, both to benchmark the efforts of countries and to provide advice on reforms to improve quality of health care. This series of individual country reviews will be followed by a final summary report on the lessons learnt for good policy practices.

ContentsAssessment and recommendations

Chapter 1. Quality of care in Turkey’s health system

Chapter 2. Primary care in Turkey

Chapter 3. Improving hospital care in Turkey

Chapter 4. Paying for health care in Turkey

http://www.oecd.org/els/health-systems/health-care-quality-reviews.htm

OE

CD

Review

s of H

ealth Care Q

uality Tu

Rk

Ey

Ra

isin

g s

Tan

Da

RD

s

OECD Reviews of Health Care Quality

TuRkEyRaising sTanDaRDs

Hulett_l

Typewritten Text

EXECUTIVE SUMMARY, ASSESSMENT AND RECOMMENDATIONS 25 November 2014

Hulett_l

Typewritten Text

Health Division Directorate for Employment, Labour and Social Affairs

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

www.oecd.org/els/health-systems/health-care-quality-reviews.htm

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

Hulett_l

Typewritten Text

www.oecd.org/els/health-systems/health-care-quality-reviews.htm

OECD Reviews of HealthCare Quality:

Turkey2014

RAISING STANDARDS

This work is published on the responsibility of the Secretary-General of the OECD.The opinions expressed and arguments employed herein do not necessarily reflectthe official views of the Organisation or of the governments of its member countries.

This document and any map included herein are without prejudice to the status ofor sovereignty over any territory, to the delimitation of international frontiers andboundaries and to the name of any territory, city or area.

ISBN 978-92-64-20204-7 (print)ISBN 978-92-64-20205-4 (PDF)

Series: OECD Reviews of Health Care QualityISSN 2227-0477 (print)ISSN 2227-0485 (online)

The statistical data for Israel are supplied by and under the responsibility of the relevantIsraeli authorities. The use of such data by the OECD is without prejudice to the status of theGolan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms ofinternational law.

Photo credits: Cover © Art Glazer/Getty Images.

Corrigenda to OECD publications may be found on line at: www.oecd.org/publishing/corrigenda.

© OECD 2014

You can copy, download or print OECD content for your own use, and you can include excerpts from OECDpublications, databases and multimedia products in your own documents, presentations, blogs, websites andteaching materials, provided that suitable acknowledgement of OECD as source and copyright owner is given.All requests for public or commercial use and translation rights should be submitted to [email protected] for permission to photocopy portions of this material for public or commercial use shall be addresseddirectly to the Copyright Clearance Center (CCC) at [email protected] or the Centre français d’exploitation dudroit de copie (CFC) at [email protected].

Please cite this publication as:OECD (2014), OECD Reviews of Health Care Quality: Turkey 2014: Raising Standards, OECDPublishing.http://dx.doi.org/10.1787/9789264202054-en

FOREWORD – 3

OECD REVIEWS OF HEALTH CARE QUALITY: TURKEY © OECD 2014

Foreword

This report is the sixth of a series of publications reviewing the quality of health care across selected OECD countries. As health costs continue to climb, policy makers increasingly face the challenge of ensuring that substantial spending on health is delivering value for money. At the same time, concerns about patients occasionally receiving poor quality health care led to demands for greater transparency and accountability. Despite this, there is still considerable uncertainty over which policies work best in delivering health care that is safe, effective and provides a good patient experience, and which quality-improvement strategies can help deliver the best care at the least cost. OECD Reviews of Health Care Quality seek to highlight and support the development of better policies to improve quality in health care, to help ensure that the substantial resources devoted to health are being used effectively in supporting people to live healthier lives.

Turkey’s Health Transformation Programme, which began in 2003, has rightly been commended for extending health insurance, increasing the supply of primary care (particularly for maternal and child health) and increasing access to hospital care. These were undoubtedly the right early priorities to choose. Now, however, a new focus is needed on the quality of health care. Increasing the amount and impact of quality-related data will be essential to this, both at service level and nationally, making data publicly available so that comparison with peers can be the basis for constant quality assurance and improvement. Payments to hospitals should be refined to better reflect the complexity of individual cases and limit a tendency to over-supply, especially where the same care could be better provided in primary or community care. The new specialty of family medicine, responsible for delivering primary care, could also be strengthened through a number of initiatives, such as patient registers, more extensive indicators of care and its outcomes, and clinical guidelines – particularly for long-term conditions such as heart disease or diabetes.

4 – ACKNOWLEDGEMENTS


ACKNOWLEDGEMENTS

This report was managed and co-ordinated by Ian Forde. The other authors of this report are Caroline Berchet, Emily Hewlett, Niek Klazinga and Ankit Kumar. The authors wish to thank Stefano Scarpetta, Mark Pearson and Francesca Colombo from the OECD Secretariat for their comments and suggestions. Thanks also go to Marlène Mohier, Nathalie Bienvenu and Lucy Hulett for their editing and to Judy Zinnemann for assistance.

The completion of this report would not have been possible without the generous support of Turkish authorities. This report has benefited from the expertise and material received from many health officials, health professionals, and health experts that the OECD review team met during a mission to Turkey in February 2013. These included officials from the Ministry of Health, from the General Directorate of Health Information Systems General Directorate, from the Strategy Development Department, the Association of Private Hospitals, the Turkish Medical Association, the Turkish Public Health Institution, the Turkish Nurses Association, the Turkish Association of Midwives, the Turkish Association of Family Physicians, Hacettepe University, the Turkish Public Hospitals Agency, the General Directorate of Health Services, the Turkish Quality Association, the Social Security Institution and the Turkish Health Care Workers Union.

The review team is especially thankful to Hakan Ari from the Ministry of Health, to Dr. Bilgehan Karadayi from the Health Technology Assessment Department at the General Directorate of Health Research at the Health Ministry, and to professor Dr. U ur Dilmen from the General Director of Health Research at the Ministry of Health for their help in setting up the mission and co-ordinating responses to a questionnaire on quality of care policies and data. The report has benefited from the invaluable comments of Turkish authorities and experts who reviewed an earlier draft.

TABLE OF CONTENTS – 5


Table of contents

Acronyms and abbreviations ................................................................................. 9 Executive summary ............................................................................................... 11 Assessment and recommendations ...................................................................... 15

Chapter 1. Quality of care in Turkey’s health system ........................................ 33

1.1. Introduction ...................................................................................................... 34 1.2. Context ............................................................................................................. 35 1.3. Profiling policies on quality of health care and their impact ........................... 40 1.4. Health system design ....................................................................................... 40 1.5. Assuring the quality of inputs to the Turkish health-care system .................... 47 1.6. Patient safety policies ...................................................................................... 51 1.7. Health system monitoring: Building an information infrastructure for measuring quality .............................................................................................. 51 1.8. Conclusions ...................................................................................................... 54 References ............................................................................................................... 57

Chapter 2. Primary care in Turkey ...................................................................... 59

2.1. The configuration of primary care in Turkey ................................................... 60 2.2. Outcomes associated with primary care ........................................................... 63 2.3. The emerging challenge of long-term conditions in Turkey ............................ 67 2.4. Building the next phase of primary care reform: Quality assurance and improvement .................................................................................................... 73 2.5. Conclusions ...................................................................................................... 84 Notes ....................................................................................................................... 86 References ............................................................................................................... 87

Chapter 3. Improving hospital care in Turkey ................................................... 89

3.1. The Turkish hospital sector has seen strong capacity development ................. 90 3.2. Strengthening quality governance in the hospital system and creating a quality culture ....................................................................................................... 99 3.3. Shifting from quality of hospital care to keeping people out of hospitals ......... 105 3.4. Conclusions .................................................................................................... 110 References ............................................................................................................. 111

6 – TABLE OF CONTENTS


Chapter 4. Paying for health care in Turkey ..................................................... 115

4.1. Context: Rising spending and fewer human resources than most OECD countries .................................................................................................... 116 4.2. There has been a major transformation in the financing of health care services .......................................................................................................... 120 4.3. Financing of hospitals .................................................................................... 125 4.4. Paying for primary care .................................................................................. 134 4.5. Conclusions .................................................................................................... 137 References ............................................................................................................. 139

Figures

Figure 1.1. Life expectancy at birth, 1970 and 2011 .............................................. 36 Figure 1.2. Case-fatality in adults aged 45 and over within 30 days after admission for AMI (admission-based), 2011 (or nearest year) ................................ 38 Figure 1.3. Case-fatality in adults aged 45 and over within 30 days after admission for ischemic stroke (admission-based), 2011 (or nearest year) ............... 39 Figure 1.4. Structure of the Ministry of Health following the 2011 re-organisation ............................................................................................... 43 Figure 2.1. Physician salaries adjusted to 2011 price levels ................................... 63 Figure 2.2. Generalists and specialists as a share of all physicians, 2011 (or nearest year) ...................................................................................................... 64 Figure 2.3. Number of GPs per 1 000 population, 2002 and 2011 ......................... 65 Figure 2.4. Decline in infant mortality rates, 1970-2011 (or nearest year) ............. 66 Figure 2.5. Vaccination rates for measles, children aged 1, 2011 (or nearest year) ...................................................................................................... 66 Figure 2.6. Trend in provision of minimum antenatal care (at least four checks), 2003-11 ..................................................................................................... 67 Figure 2.7. Distribution of provision of minimum antenatal care (at least four checks), 2011 ..................................................................................... 67 Figure 2.8. Trend in infant follow-up visits, 2000-11 .............................................. 67 Figure 2.9. Distribution of infant follow-ups, 2011 ................................................. 67 Figure 2.10. Prevalence of obesity among adults, 2011 (or nearest year) ............... 69 Figure 2.11. Prevalence of females and males smoking daily, 2011 (or nearest year) ...................................................................................................... 69 Figure 2.12. Self-reported COPD by highest attained level of education, 2008 (or nearest year) ............................................................................................. 70 Figure 2.13. Asthma hospital admission rates, 2011 (or nearest year) ................... 72 Figure 2.14. Structure of the Quality Indicators in Community Healthcare (QICH) programme, Israel .................................................................... 76 Figure 3.1. Turkey has the third lowest number of hospital beds in the OECD ............................................................................................................ 91

TABLE OF CONTENTS – 7


Figure 3.2. The number of hospitals in Turkey is a third lower than the OECD average .................................................................................................. 91 Figure 3.3. The number of hospital beds in Turkey has been growing steadily .................................................................................................................... 92 Figure 3.4. Bed occupancy in Turkey remains the second lowest in the OECD ............................................................................................................ 94 Figure 3.5. Hospital discharges in Turkey have doubled since 2000 ...................... 95 Figure 3.6. Caesarean sections in Turkey are the highest and faster growing in the OECD ............................................................................................................ 97 Figure 3.7. Expenditure on hospitals in Turkey has been among the fastest in the OECD ............................................................................................................ 99 Figure 4.1. Health expenditure per capita, 2011 (or nearest year) ......................... 117 Figure 4.2. Health spending per capita by provider, 2003 to 2008 ....................... 118 Figure 4.3. Doctors per 1 000 population, 2000 and 2011 (or earliest available) ............................................................................................. 119 Figure 4.4. Ratio of nurses to doctors, 2011 (or earliest available) ...................... 119 Figure 4.5. Change in out-of-pocket spending ...................................................... 124 Figure 4.6. Evolution of remuneration for Ministry of Health personnel ............. 132 Figure 4.7. Hospital discharges per 1 000 people, Turkey and OECD average for 2000, 2004 and 2010 .......................................................................... 133

Tables

Table 1.1. A typology of health-care policies that influence health care quality .............................................................................................................. 40 Table 3.1. Growth in the public and private hospital sector in Turkey ................... 93 Table 3.2. Turkey gatekeeping rules are weak ...................................................... 107

ACRONYMS AND ABBREVIATIONS – 9


Acronyms and abbreviations

AHRQ Agency for Healthcare Research and Quality AMI Acute myocardial infarction CHD Coronary heart disease CME Continuous Medical Education COPD Chronic obstructive pulmonary disease (chronic bronchitis

and emphysema) CPD Continuing professional development CT Computed tomography DALY Disability-adjusted life year DRG Diagnostic-Related Group EFQM European Foundation of Quality Management FM Family medicine FP Family physician GDP Gross domestic product GHIS Genel Sa lık Sigortası (General Health Insurance Scheme) GP General practitioner HAS Haute Autorité de Santé HCQI OECD Health Care Quality Indicator Programme HTP Health Transformation Programme ISQua International Society for Quality in Health Care IT Information technology KETEM Cancer Early Diagnosis Screening and Training Centres LTC Long-term care MCH Maternal and child health MDT Multi-disciplinary team MRI Magnetic resonance imaging NCD Non-communicable disease

10 – ACRONYMS AND ABBREVIATIONS


NCHOD National Centre for Health Outcome Development NHS National Health System NICE National Institute for Clinical Excellence NSQHS National Safety and Quality Health Service Standards PCP Primary care physician PTCA Percutaneous transluminal coronary angioplasty QICH Quality Indicators in Community Healthcare RHA Regional Hospitalisation Agency SSI Sosyal Güvenlik Kurumu (Social Security Institution) TAHUD Türkiye Aile Hekimleri Uzmanlık Derne i (Turkish

Association of Family Physicians) TRL Turkish lira TUIK Turkish Statistical Institute TURDEP Turkish Diabetes Epidemiological

EXECUTIVE SUMMARY – 11


Executive summary

Over the past decade, Turkey has implemented remarkable health-care reforms, achieving universal health coverage in 2003, and dramatically expanding access to care for the population. Accompanied by significant investment in the hospital sector and the establishment of a family physician system, the Health Transformation Programme (HTP) has delivered a high level of activity in the health system. The reforms benefitted from ambitious leadership and a clear set of priorities (focused on expanding health insurance and improving access and, in the clinical domain, on maternal and child health). An evaluation culture built in from the beginning and a willingness to open up the reform process to external scrutiny were also fundamental elements. Centralisation and rationalisation of the health system’s governance was critical in achieving recent health-care successes. A maturing system, however, might now benefit from a less directive approach. Indeed, the centre should now feel confident enough to relax control, and instead set out the broad ambitions and get the right incentives in place, focussing on a quality governance role. The Ministry of Health is, it should be noted, taking some steps to devolve responsibility for providing hospital services and focus instead on its regulatory, oversight, and quality governance functions.

Routinely published data in the Turkish health system largely focuses on supply and activity, hence there is great scope to increase the role of quality-related data in steering Turkey’s health system. Data on health sector activity and outcomes also need to be made more available and more usable for individual patients and clinicians. Open comparison of service-level data should be promoted, with the expectation that it will function as a highly effective tool to drive up quality standards and reduce variation. Strengthening of the involvement of all stakeholders in the standard setting process and increased transparency on the process of evaluation and scoring can help to further increase the acceptability and impact of the reforms. National statistics must also start collecting the right data in the right format to allow Turkey to participate in international benchmarking activities, such as the OECD’s health care quality indicators. Continued work on specific registries, whilst optimising the use of routine administrative data in

12 – EXECUTIVE SUMMARY


tracking and improving the quality of services, is also needed. A coherent policy on strengthening the Turkey’s health information infrastructure to facilitate the use of quality indicators is needed.

Whilst focussing on coverage, access and activity were undoubtedly the right priorities in the first decade of the HTP, Turkey’s health system must now focus on quality and outcomes. A clear example of the risks of not doing so comes from the hospital sector. With a high propensity for visiting emergency hospital services even for minor ailments, payment incentives that encourage volumes of in-patient care, and initiatives to improve clinical quality of care in early stages, there is a risk that a focus on quantity and productivity enhancement might come at the expense of ensuring that care is delivered in the most appropriate care settings. In parallel, the primary care sector, having achieved of widely hailed improvements in access, must now adopt quality as the focus of on-going reform. Turkey has a number of initiatives in place with the potential to be effective tools for quality assurance and improvement, but each must be developed further before their full utility can be exploited. Current quality assurance activities tend to focus on minimum standards, and limited information is available in the public domain. To build a quality culture, the focus of on-going reform should move from one of control and penalising bad performers to one of encouraging continuous improvement.

Payments to hospitals – particularly public hospitals – have undergone a major transformation over the course of the past decade, but remain activity-focused and poorly linked to outcomes. The quality component that does exist in hospital reimbursement is weak (and self-assessed). Furthermore, there is little incentive for public hospitals to contain costs, since overspends are met with increases in the global budget the following year. Neither is there any differentiation in reimbursement between tertiary centres taking the most complex cases and general hospitals. This situation could be addressed through the gradual shift from package-fees to a case-mix adjusted payment. Turkey already has the data infrastructure and coding processes in place in that would facilitate the shift to case-mix adjustment relatively rapidly. Public hospitals pay staff through one of the largest pay for performance schemes among OECD countries, although the majority of these are indicators of productivity measured by health outputs. Similarly, in primary care, the quality component of physicians’ pay is, in fact, activity based.

Turkey’s success at improving health-care coverage and system performance has been impressive, and key areas of the reform are reflected in the significant improvements across indicators such as maternal mortality, and infant mortality. Although maternal and child health were undoubtedly the right investments to have made in the early years of Turkey’s HTP,

EXECUTIVE SUMMARY – 13


Turkey’s maturing health system must anticipate the inevitable shifting of the national disease burden toward the chronic morbidities associated with increasing years and, in some cases, unhealthy lifestyles. This transition, coupled with increasing patient expectations around more convenient and better co-ordinated care, ought to renew the focus on primary care.

ASSESSMENT AND RECOMMENDATIONS – 15


Assessment and recommendations

Driven by clear vision and strong leadership, the first ten years of Turkey’s Health Transformation Programme have dramatically expanded access to health care. Accompanied by significant investment in the hospital sector, the establishment of a family physician system and payment reforms, the Health Transformation Programme (HTP) has delivered better access and a high level of activity in the health system. To build on its success, it is time for Turkey to shift the emphasis from encouraging high volume of care to delivering high-quality health services.

The Turkish HTP demonstrates how a country can, in a relatively short period of time, successfully deliver universal health coverage. The reform consolidated multiple coverage schemes that had diverse entitlement rules into a single Social Security Institution (SSI), improving pooling and redistribution. The Green Card programme for the poor, the main social protection programme of the Turkish Government, was the last to be added to the Social Security Institution in 2012. Expansion in coverage was accompanied by health spending growth well above other OECD countries, averaging 7.7% since 2002. As a share of GDP, health spending in Turkey went from 5.4% in 2002 to 6.1% in 2008.

Re-building a primary care system to maintain and improve population health – with clearly assigned responsibilities for service delivery and nationally consistent payment methods – has been a central ambition of the reform. The training of new physicians as family physicians and retraining of existing GPs into the speciality have been national priorities. The introduction of a pay penalty for doctors not delivering a basic set of child and maternal health services led to remarkable improvements in the delivery of these services. Major investments in hospital capacity across the public (Ministry of Health) and the private sector, and the introduction of payment reforms that linked the remuneration of specialists in public hospitals to the volume of services delivered, have also increased service delivery in secondary care. The private sector has developed rapidly over the past decade, including in less developed regions of Turkey, and now represent 36% of hospitals and nearly 18% of all hospitals beds.

16 – ASSESSMENT AND RECOMMENDATIONS


In a sector where reforms are difficult to design and even harder to deliver, the Turkish Government deserves praise for the way in which the health care reform agenda has progressed. Central governance and ambitious leadership have been instrumental to the implementation of this reform programme. The HTP also benefitted from a clear set of priorities – focused on expanding health insurance and improving access and, in the clinical domain, on maternal and child health – as well as external advice and support from international agencies. Particularly important was a willingness to open up the reform process to external scrutiny from outside the country.

While still too early to be fully evaluated, the reforms have undoubtedly been a success in several respects. The health of the Turkish population has improved impressively over the past ten years as illustrated by statistics on life expectancy at birth, neonatal mortality, maternal mortality and infant mortality. For example, life expectancy at birth has reached 74 years, recording the second largest gain in the OECD – 25 years since 1960. Similarly, Turkey has achieved the highest average reduction of 6.9% per year in infant mortality between 1970 and 2011, followed in the OECD by Korea (6.4% per year) and Portugal (6.8% per year). Financial protection has also greatly improved. Public health spending now accounts for 73% of total health expenditure, slightly above the OECD average of 72%. Over the past decade, out-of-pocket spending by families has shown the second fastest reduction after Korea, and reported figures are now the lowest in the OECD as a share of household consumption (1.5%). Likewise health-care facilities and infrastructure have expanded, coming closer to OECD averages.

The reform agenda is not over yet. Although capacity in the Turkish hospital sector has been growing fast, there are still major plans for building new and modernising existing hospital facilities in the Ministry of Health sector. There are also ambitions to continue expanding the primary care workforce and achieve by 2023, in time for the centenary celebrations of the founding of the Turkish Republic, a doctor to population ratio that matches the norm amongst OECD countries.

However, despite universal health coverage, health indicators still remain among the lowest in the OECD, and a number of challenges remain in the Turkish health system:



• Having placed much emphasis on improving productivity in health care, the system should now fully embrace quality and outcome improvement as the next overarching priority.

• Overly centralised governance of the health system could stifle local-level initiative and flexibility. Although instrumental to delivering the significant reforms Turkey has pursued in the past decade, there is a risk that heavy centralisation could discourage constructive involvement and initiatives from providers, in a country whose geography, epidemiology of disease and ethnicity are highly diverse.

• Few indicators of quality are collected, and those that exist, point to poor quality of care by OECD standards. For example, mortality within 30 days of hospital admission for acute myocardial infarction in Turkey – 10.7 per 100 patients – is 35% higher than the OECD average of 7.9. Similarly for stroke, case fatality within 30 days of hospital admission is the third highest in the OECD (11.8 per 100 patients), following Mexico and Slovenia. These data signal the need for prioritising monitoring and improvement initiatives. There is still insufficient collection and public reporting of quality measures, including from the private sector.

• Dialogue between key stakeholders has not always been constructive, while professional efforts to pursue modern form of continuous medical education are still in their infancy. This might mitigate efforts to further drive quality gains in health-care services.

• Payment systems have rewarded structure and activity very well; however the link between increased activity and quality of care can in no way be assumed, and there is a risk that productivity might lead to higher cost without necessarily improving outcomes. The dependency of public hospitals for funding from the Ministry of Health risks central government taking more interest in the operations of its own facilities rather than in assuring quality for the system as a whole.

• Emerging new health care needs will challenge the health system in its current configuration. While maternal and child health have rightly been the priorities for the Turkish health system in the past decade, fast economic growth and reductions in premature mortality mean that Turkey will face a demographic and epidemiological shift at a much faster speed than most OECD countries. Chronic diseases such as diabetes and risk factors such as obesity must urgently become a focus of policy makers and clinicians’ attention.



Having demonstrated remarkable confidence in pushing reform, the Turkish health-care system is very well placed to address the challenges highlighted above, maturing into a system that is adaptable and ready to address emerging health needs. Whilst focussing on coverage, access and activity were undoubtedly the right priorities in the first decade of the HTP, Turkey’s health system must now focus on quality and outcomes. There are several opportunities for doing so, each implying further reforms, as set out in the text that follows.

Further efforts are needed to place focus of health policy on outcomes and quality

The focus of on-going reform should move from quantity to quality assurance

Compared to other OECD countries, there is less evidence of a quality culture in Turkish health care than elsewhere. Thus far, reforms have had a near-exclusive focus on inputs and activity, with a view to increasing the volumes of both.

These, however, are only a part of what guarantees quality in health care, and the Turkish authorities and health professionals need to start focusing on outcomes. The current bias toward supply and activity, and relatively weak quality culture that ensues as consequence, is evident in several examples:

• In primary care, services are inspected against standards that focus almost exclusively on the physical fabric of the building and availability of clinical equipment and emergency drugs. A random sample of 10% patient records is examined regularly. A limited number of activity-related standards are included, which focus on maternal and child health. These are binary measures whether all babies, for example, have had a new-born hearing test. No standards relate to outcomes.

• In the hospital sector, patients’ propensity for visiting emergency hospital services even for minor ailments and payment incentives that encourage volumes of in-patient care create the risk that patient safety and care effectiveness are not prioritised enough.

• Although an adverse event reporting system for public hospitals has been established, reporting is currently voluntary. The extent to which such a system can support hospitals to identify common adverse events and learn means to avoid them is therefore limited.



• The narrative of co-ordination and integration, which is an increasing priority in many OECD health systems, is nearly absent in Turkey. However, some progress has been made in this direction through improving health information systems to foster communication between health sectors.

…and to building professional interest in quality as well as a culture of quality improvement

Whilst Turkey has successfully expanded the number of health professionals and improved their distribution, the country still has fewer doctors relative to its population (1.7 per 1 000 population) than other OECD countries (3.2 per 1 000 population, on average across the OECD). Furthermore, the focus thus far has been on numbers – attention to quality has been allowed to lapse. In the case of family medicine, the original two-year retraining programme for the cohort of pre-HTP primary care physicians is often curtailed in an effort to get as many to qualify in the new speciality as possible. Apart from the new speciality of family physicians (described in Chapter 2), Turkey currently has no formal guidance, or requirements at national level on continuing professional development. Some local initiatives are conducted by the Ministry of Health or professional associations, including conferences, symposiums and post-graduate courses to train physicians, nurses, technicians and other health-care workers. Turkey needs a balanced system of self-regulation and accountability of the clinical professions in order to assure quality of care as in most other OECD countries.

To build a quality culture, the focus of on-going reform should move from one of control and penalising bad performers to one of encouraging continuous improvement. Clinicians and service managers should be encouraged to change practice towards better and safer care through a mix of educational measures, data collection and disclosure requirements with feedback on performance provided back to clinicians, managers and users. The celebration of good practices or encouragement of hospital and clinician “champion roles” will also contribute to a quality improvement culture.

There is also a need to monitor the outcomes of recent reforms that regulate the extent of private practice of hospital physicians. Whilst this reform is likely to have protected patients from being referred unnecessarily to physicians’ private practice, or from being seen by exclusively by junior doctors in public hospitals, there are reports that it has led to some clinicians dropping part-time practice in public hospitals. A review of quality and access indicators before and after the reform should be undertaken.



Information can be better exploited to steer improvement in Turkey’s health system

Data systems on quality of care are still under development in Turkey. Although a growing amount of data is becoming available that can be used to monitor quality of care, present quality assurance activities are rather control-oriented and the available information is not exploited to its full potential. Very limited information on quality is available in the public domain and focuses on supply and activity – such as consultation numbers, hospital discharge rates and lengths of stay and staff remuneration. Clinical outcomes of care – apart from very broad societal measures such as life expectancy – are not routinely reported. Furthermore, what data is viewable to service providers or users is presented at a high aggregate level, which can confound local efforts to benchmark and monitor quality improvement.

Among the priorities for improving the information system on quality for Turkey, the following seem appropriate:

• Although a growing amount of data is available to monitor quality of care in public hospitals, further development is needed over the coming years to strengthen the collection and reporting of a broad set of quality indicators, particularly in areas beyond maternal and child health, such as non-communicable diseases or mental health.

• Performance measurement efforts developed by the Ministry of Health for public hospitals can be furthered. In particular, strengthening of the involvement of all stakeholders in the standard-setting process and transparency on the process of evaluation and scoring can help to further increase the acceptability and impact of the programme.

• A coherent policy on how to strengthen the Turkish information infrastructure to facilitate the use of quality indicators, addressing topics such as data-linkage, secondary use of data from Electronic Health Records and assurance of privacy and data-security is advisable. Turkey could look at the experience of other OECD countries that have made significant progress on these issues, such as South Korea, Finland, Sweden and the United Kingdom.

• Further work on specific registries should be encouraged as well as better use of administrative data available through organisations such as the Social Security Institute. A national cancer registry, for example, would be a natural evolution of the network of KETEMs, or early diagnosis centres, that Turkey has established and allow the patterns and outcomes of cancer care to be more closely scrutinised and opportunities for improvement identified. This data



development work can be linked to developing a more sophisticated set of standards, focussing on the processes and outcomes of clinical care.

• Data on health sector activity and outcomes needs to be made more available and usable for patients and clinicians. From the user perspective, steps have been taken to strengthen the position of patients around complaint handling. It is advisable now to have more information on performance of health-care services in the public domain. Capturing the experiences of health-care users systematically could be more broadly embedded. Open comparison of service-level data across different provinces as in Sweden should be promoted, with the expectation that it would function as a highly effective tool to drive up quality standards and reduce variation across regions and providers.

• National statistics must start collecting the right data in the right format to allow Turkey to participate in international benchmarking activities, such as the OECD Health Care Quality Indicators project, including systematic measurement of patients’ experiences of using the health-care system. It is expected that these statistics such as five-year survival rates for cancer, 30-day case fatality rates for patients admitted for AMI and stroke and hospital admission rates for quality of ambulatory care sensitive conditions such as diabetes, chronic heart failure and COPD will become available over the coming years. But significant extra investment is needed to ensure that these are robust enough to submit for international comparison.

Centralisation of the health system’s governance has been a critical element in achieving success but should now be relaxed

Centralisation is a dominant feature of the Turkish governance model, both geographically (there is limited local autonomy) and functionally (through managing delivery of key health system activities from within the Ministry of Health). One of the benefits of such strong centralisation has been the ability to prioritise and rapidly roll-out key health system functions that were previously weak or variable, for example in the primary care sector and on payment arrangements.

However, the heavy centralisation of power which has characterised the first decade of Turkey’s HTP engenders a number of trade-offs, such as a lack of innovation and diffusion of knowledge in the hospital sector and few local-level incentives or opportunities for workforce development. Limited flexibility in per capita payments to doctors, which are fixed centrally, has also created challenges in supporting doctors who wish to employ other



health professionals and deliver team-based care. Family physicians, for example, have no budgetary oversight over the nurses working under their supervision.

Another important consequence has been conflict with professional bodies. Whilst strongly divergent views between government and professional groups are not unique to Turkey, consensus has had a notably minor role in steering reform in Turkey. The Turkish Medical Association remains opposed to the HTP, especially as payment arrangements place considerable pressure on doctors to deliver higher volumes of services. Although some elementary self-regulation of the medical profession is in place, strong disagreements exist between the profession (as represented by the Turkish Medical Association, Turkish Nurses’ Association and Turkish Midwives’ Association) and the government on mutual roles and responsibilities.

Having achieved impressive reforms, the central government authorities should now feel confident enough to relax functional and operational control. While the Ministry of Health is taking some such steps to devolve responsibility, for example by devolving the responsibility for delivering hospital services, progress on shifting focus to its regulatory, oversight, and quality governance functions, has been rather slow. To respond to the challenges that Turkey is now facing, government authorities might more usefully redefine its function as one of setting the broad goals for the system and ensuring that the right incentives are in place. Releasing responsibility for operations will also free central government authorities to focus on a quality governance role.

Reforms in the primary health care represent an excellent platform for further quality improvement efforts

One of the Health Transformation Programme’s central ambitions was to rebuild primary care

Whereas prior to 2003 arrangements for primary care were only loosely defined, with a doctor and/or ancillary staff such as nurses and midwives offering a variable range of services to a locality, often dependent on individual initiative, the HTP established a family medicine system in 2005 to bring consistency and structure to the sector. The core team was defined as a family physician and a nurse and made responsible for a core set of tasks, focussed around maternal and child health. Original plans envisaged that existing GPs would progressively gain recognition as family physicians provided they completed ten days’ of preliminary orientation, followed by a two-year programme of specialist training.



The efforts have borne fruit. Both the absolute numbers of primary care physicians and their distribution has dramatically improved since the implementation of the HTP. Between 2000 and 2008, the primary care workforce expanded from 41.1 doctors per 100 000 to 52.6, and the ratio between the best and least-served areas improved from 8.3:1 to 2.8:1. Turkey’s primary care/generalist workforce now comprises 33% of all doctors, in line with the OECD average of 30%.

Discrete incentivised activities can be linked to better outcomes, but the broader picture of primary care quality is much less clear

An important aspect of the reform has been the change in payment mechanisms. Family physicians are reimbursed by prorated capitation payments alongside fees-for-service. To encourage delivery of some of the key antenatal and postnatal care, the payment system embeds an element of performance-related pay, by applying a penalty of around USD 220 to physicians failing to offer, for example, breastfeeding and contraceptive advice, or growth and development monitoring and immunisation for children up to two years of age. The programme has been successful in improving maternal and child health, perhaps the central aim of the HTP. Besides the already mentioned data on child mortality and vaccination, the proportion of women who have attended at least four prenatal visits rose from 53.9% in 2003 to 73.7% in 2008 and the proportion of births attended by skilled health staff rose from 83% to 91.3% over the same period, in line with the anticipated effects of the incentive schemes.

Yet, beyond maternal and child health, the quality of primary care (whether measured in terms of activity or outcomes) is much less clear. For example:

• Cancer screening rates, a core primary care activity, are low. In 2011, only 15.5% of Turkish women aged 20-69 were screened for cervical cancer, compared to an OECD average of 59.6%. 27.3% of women aged 50-69 were screened for breast cancer (OECD average 61.5%) and 3.2% of adults aged 50-74 were screened for colorectal cancer (EU15 average 12.7%).

• Important measures of quality of primary care that are collected by other OECD counties are not available for Turkey as yet, although work is underway to develop them. This is the case for example for rates of hospital admission for chronic conditions deemed fully manageable within primary care, such as asthma, chronic obstructive pulmonary disease (COPD) or diabetes. Other relevant measures, such as the rate of lower limb amputation or frequency of annual retinal exam in diabetics, are not available either.



Towards quality assurance and quality improvement in primary care Having achieved widely hailed improvements in access to family

medicine, and with a programme in place for its continued expansion, the quality of primary care must now become the focus of on-going reform. Three main priorities stand out.

The first will be to strengthen indicators of primary care activity and outcomes, and, especially the feedback loop back to professionals. Currently, family physicians are required to return data on maternal and child health to the Health Information Systems Directorate at the Health Ministry, through the Sa lık Net platform (“Health Net”). Some prescribing data is also routinely collected. Activity across antenatal care, and childhood vaccination is visible to the Health Information Directorate at regional, institutional and individual-practitioner level, but that fact that this is not returned to the institution or practitioner with relevant peer-comparisons is a missed opportunity for quality improvement. Turkey could look at other OECD countries experiences that have introduced system of annual appraisal and feedback to clinicians, such as the United Kingdom, and developed sophisticated monitoring of quality in primary care, such as Israel.

The second priority concerns standards. As mentioned earlier, a limited number of activity-related standards on maternal and child health are collected. What needs to be developed is a more sophisticated set of standards focussing on the processes and outcomes of clinical care. Turkey has embarked on an ambitious programme to translate an extensive set of clinical guidelines written by the Finnish Medical Society but the difficulties of embedding a large number of guidelines at once, and in particular of changing practice through guidelines with little sense of local ownership or participation during development, should not be underestimated. It may be more effective and instructive for Turkish stakeholders to choose a priority clinical area – cardiovascular disease or diabetes would be obvious choices – and develop home-grown management guidelines for local implementation.

Third, thought should be given to future expansion of the role of primary care, particularly in anticipation of an epidemiological shift and rising burden of long-term conditions and multi-morbidity. Although maternal and child health were undoubtedly the right initial investments to have made in the early years of Turkey’s HTP, a mature primary care service needs to make a comprehensive offer and be the trusted first point of contact for the vast majority of health needs, irrespective of age or gender. In particular, Turkey’s maturing health system must anticipate the inevitable shifting of the national disease burden toward the chronic morbidities associated with increasing years and, in some cases, unhealthy life styles.



This transition, coupled with increasing patient expectations around more convenient and better co-ordinated care, will require increasing the robustness of Turkey’s information systems, as well as reinforcing the hierarchy between levels of care, to avoid unnecessary use of secondary care.

Delivering quality gains from growth in hospital capacity and utilisation

Fast growth in hospital capacity and activity could raise questions about quality

Contrary to efforts in most OECD countries to downsize the hospital sector, capacity in the Turkish hospital sector has been growing fast. The increase in the number of hospital beds since 2000 has been second only to that of Korea. The number of hospitals in Turkey has increased from 1 153 in 2002 to 1 453 in 2011. There are major plans for building new and modernising existing hospital facilities in the Ministry of Health sector, and incentives have stimulated the development of a significant private hospital sector, which have helped to fill in gaps in capacity in several regions.

Nevertheless, some statistics suggest that the way the hospital sector has been developed might pose challenges in the absence of an appropriate quality governance framework. For example:

• Bed occupancy rate in Turkish hospitals is only 64.9%, the third lowest in the OECD after the Netherlands and the United States, and it is only just above 50% in the private hospital sector. These figures might indicate either low demand, or capacity in excess of need. The number of hospital discharges, while still lower than in two-thirds of OECD countries, has doubled since 2000, the second fastest rate of growth in Europe. Although there is a lack of data around unnecessary hospitalisations, this increase is probably due to better access to care but could also be related to unnecessary hospitalisation.

• While few data on procedures are available, data on the number of caesarean sections in Turkey are the highest in the OECD. Between 2006 and 2011 the number of caesarean sections increased from 297 to 462 per 1 000 live births, which is the highest rate of growth in the OECD. A high rate of caesarean section can increase the risk of mortality and morbidities for both the mother and the child, or lead to risk of complications for future deliveries.



• The few data collected on acute care – such as mortality after 30 days of hospitalisation for acute myocardial infarction or for ischemic stroke – point to rates well above the OECD average.

Although some Turkish hospitals have started to reduce the number of beds, and the private sector is not permitted to increase its bed capacity, these data suggest that Turkey might need to monitor its hospital sector more closely. In particular, there is a need to understand whether and how current trajectories of activity and capacity are leading to improvement in quality of care. Data infrastructure and monitoring of quality of care in all of Turkey’s hospitals also needs to be strengthened and the underpinning data infrastructure developed across the public and private sector. Currently, while private hospitals claim that they collect already some quality indicators, few of these are made available either to the Ministry of Health or to patients.

Strengthening quality governance must occur uniformly across the whole secondary care system

Turkey should look to strengthen quality governance (standards, monitoring and transparency) in the hospital system as a whole, holding both private and public sector hospitals to the same high standards of delivering effective, safe and patient-centered care. Experience from other OECD countries (Australia, United States, France, England) shows how governments have a key role to play in creating an even playing field across different hospital sectors through, for example, introducing third-party accreditation in addition to government accreditation of hospitals and specifying minimum standards across the public and private sectors.

An impressive number of initiatives has been taken by the Department of Health Care Quality and Accreditation, (2 226 institutions were evaluated for their quality in 2012). However, focus has been so far on structural and organisational components of hospitals. Broadening the programme towards clinical outcomes of health-care services is advisable to make the model more useful for formative functions such as quality improvement initiatives.

University hospitals report that they have been struggling with the impact of the HTP. Despite receiving higher payments from the Social Security Institution, they claim that special functions, most notably care for the most complex or demanding cases, are not adequately remunerated. Compounding the challenge, the regulation forbidding clinicians from working part-time across different sectors (public, private and hospitals) has meant university hospitals are conscious of the risk of losing staff (generally the most qualified doctors in the country) to the private or the Ministry of Health sector, both of which are perceived to offer better remuneration for



the case load undertaken. Other OECD countries have specific payment arrangements for university hospitals to ensure maintenance of certain public health functions. For example, France makes extra payments for teaching, research and innovation, emergencies, psychiatry, and certain rehabilitation services, while Germany has refined the structure of case-based payments to reward these functions.

Despite significant progress, challenges remain with regards to aligning public hospital and public health system governance on local and provincial levels

The Ministry of Health is pursuing efforts to devolve responsibility for providing hospital services and focus instead on its regulatory, oversight, and quality governance functions. While recent reforms to the ministry structure have set the ground for this change in functions, certain system characteristics make it more difficult for the government to take on its quality governance function. Some have already been mentioned – such as culture of centralised decision making, limited professional interest in taking on strong responsibility for quality assurance, strong financial dependency of public hospitals for funding from the ministry and insufficient public reporting of quality measures.

The governance of public hospitals, via 87 Hospital Unions, run separately from the regional population-based public health governance, may pose challenges to the co-ordination of care between the primary and secondary care sector. Care should be taken to ensure that effective dialogue takes place to ensure that secondary care services are matched to local population health needs. Furthermore, in addition to recently created affiliated agencies – such as the Public Health Institution of Turkey or the Public Hospital Institution of Turkey – it might be considered to position functions such as accreditation and health technology assessment more distant from the Ministry of Health as is the case in many other OECD countries. This is especially the case when quality governance activities apply to public health-care services as well as private health-care services.

Last, there is a strong argument to shift the locus of governance away from the central government to closer to where care is provided. Greater autonomy should be granted to provincial governments, the recently created hospital unions, and hospitals themselves to manage secondary care services.



Developing payment systems that drive higher quality care

Turkey has used payment reforms as a key instrument to achieve several of its reform targets. Both in the primary care and in the hospital sector, clear incentives for delivering more services have been built into the remuneration of providers. The remarkable results that this has delivered signal how Turkey could use payment incentives very effectively to deliver quality-related goals.

Payment systems in primary care could be better designed to reward quality

In the primary care sector, financing is in the process of shifting from funding community-based clinics to funding a new workforce of family physicians on a per capita basis. Under the system, each family medicine specialist in Turkey is paid:

• A risk-adjusted per capita payment (on average around TRL 5 600 a month or EUR 2 445), which can vary by a factor of up to 50% reflecting age, pregnant women, prisoners and other socio-economic indicators.

• 20% of monthly payment can be deducted from the per capita payment, should the family physicians fail to offer a basic set of antenatal, baby and child health services and vaccinations to at least 90% of the new children within their population.

While the quality component family physicians’ pay is activity-based, there is potential for Turkey to use the structures already in place to reward clinical outcomes. OECD primary care systems are becoming increasingly ambitious and sophisticated in designing primary care payment systems around this principle, and Turkey is in a good position with its nascent and evolving speciality of family medicine to devise some locally appropriate outcome-based incentives. Thought should also be given to expanding the current penalties to delivery of care of non-communicable diseases, for example to ensure that family physicians monitor HbA1C levels in diabetic patients.

Remaining to be addressed, however, is a concern that the payment system might, in a context of shortage of physicians, divert doctors action away from quality. Per capita payments are currently based on a catchment population of some 3 500 to 4 000 patients per doctor. This is a very large ratio of patients to doctors and may divert activities towards registration and child and maternal health care at the expense of care co-ordination, life style modification and other basic health care that can effectively be delivered in primary care. The latter services are remunerated through modest fee-for-



service payments relative to the potential for large losses in per capita base salaries. Efforts are being made, however, to increase the number of primary care doctors and reduce the list size for each doctor from current levels.

In the hospital sector, current payment and incentive structures reward neither quality nor cost-control

Payments to public hospitals have undergone a major transformation over the course of the past decade. Each public hospital service is paid by “package rates” that bundle prices for outpatient and inpatient services established by the Social Security Institute, within a global budget that is negotiated between the Social Security Institution (SSI) and the Ministry of Health. The payment of staff involves a salary and a performance-based component. Specifically, the ministry is able to distribute up to 40% of its annual budget to its medical staff in the form of performance-based payments, adjusted on the basis of a hospital specific performance score. As a result, hospital staff have a personal incentive linked to contributing to a high institutional score. This score is calculated on some 49 indicators of clinical activity, clinical processes and institutional characteristics, such as the number of invasive procedures per physician, cleanliness of hospitals, bed occupancy, average length of stay, and the share of doctors working full time. Most of these are measures of supply and activity, hence hospital incentives are poorly linked to quality outcomes (patient satisfaction rates and hospital infection rates are exceptions). As to private hospitals, they only receive a flat fee from the Social Security Institution for every patient they see and are free to charge patients additional costs, up to a ceiling fixed at 90% more than the public hospital price.

There is, in theory, an opportunity for quality-related competition amongst multiple secondary care providers in the Turkish health-care system. This is dampened, however, by the fact that providers have little incentive to offer the full range of secondary care, particularly complex packages of care for patients with the greatest need, since the current set of reimbursement codes do not reflect case severity. Furthermore, there is little incentive for public hospitals to contain costs, since overspends due to unbudgeted activity may, in some cases, be compensated with funds from other public hospitals in surplus. Prices paid by the Social Security Institution, however, are well controlled and show no evidence of inflation. As to private hospitals, publicly funded patients accessing care in private hospitals must make significant out-of-pocket contributions to meet the costs not covered by public insurance. Hence, while in theory all Turkish people are able to visit a private hospital for a service covered in the health insurance benefit package, in practice, access to private hospitals and full



exploitation of user choice is still the domain of those who can afford to pay significant out-of-pocket costs.

This situation should be addressed through the gradual shift from package fees to a case-mix adjusted payment, as most other OECD countries do. Turkey has an opportunity to use the rich data infrastructure and coding processes already in place to shift to case-mix adjustment relatively rapidly, thereby appropriately remunerating complexity of care. This should be accompanied by monitoring of quality of care for example to ensure that any expansion in numbers treated is clinically appropriate and to reduce unwarranted variation in medical practice across providers or geographical regions. This would have the further advantage of helping the Ministry of Health move fully and expeditiously in the aspired direction of focusing on quality governance of the health system and relinquishing responsibility over operations.

Policy recommendations for improving the quality of the health-care system in Turkey

Turkey Health Transformation Programme has driven groundbreaking improvement in the delivery of maternal and child health, the development of capacity and the expansion of health coverage and services. The challenge now for the Turkish health-care system will be to make quality and outcome monitoring and improvement the next overarching priority for health policy. This will require reforms to:

1. Shift the focus of on-going reform to quality assurance of health service

• Continue efforts to routinely collect and report information on the quality of care (for example in primary care, non-communicable diseases and mental health) and develop the data infrastructure for quality (for example, cancer registries and secondary use of data from electronic health records); furthermore, provide health professionals with feedback on quality measures at regional and provider level, with a view to encourage self-assessment and continuous improvement.

• Strengthen providers focus on safety and effectiveness of care for example by pursuing efforts to develop standards and requirements for continuing professional education, retraining existing primary care doctors into family physicians, and encouraging modern forms of professional self-regulation alongside accountability of the medical profession.

• Advance efforts to devolve responsibility for operations to arm’s length institutions such as the newly created Hospital Agency, the Public Health Institution of Turkey and the Pharmaceutical and Medical Devises Institution of Turkey.

• Strengthening the patient perspective by encouraging a comprehensive national adverse events reporting system and encouraging gradual diffusion of information on quality among in the public.



Policy recommendations for improving the quality of the health-care system in Turkey (cont.)

2. Encourage quality improvement efforts in the primary care sector

• Strengthen indicators on primary care outcomes that relate to the quality of primary care beyond maternal and child health care (for example on avoidable hospital admissions), and provide family physicians with access to information from the Sa lik Net platform, with relevant peer information to facilitate performance improvement.

• Pursue efforts to develop standards in primary health care related to outcomes and clinical quality of care, and focus the development of clinical guidelines on key priority areas such has cardiovascular diseases or diabetes.

• Encouraging health professionals, trough payment systems, educational measures or other means, to shift focus from maternal and child health solely to chronic diseases such as diabetes and cancer, and risk factors such as obesity.

3. Harness quality from the hospital sector

• Further efforts to collect data on safety and care effectiveness in the hospital sector (e.g., outcomes, readmission rates, complication rates) in order to monitor that expansion in hospital capacity and activity is accompanied by improvement in quality, and continue to pursue initiatives to strengthen the robustness of these data for international comparisons.

• Strengthen quality governance for the hospital system as a whole (including the private hospital sector), by extending standards developed in public hospitals to measures of hospital safety and effectiveness of care, encouraging the development of minimum standards across the public and private sector as well as third-party accreditation of hospitals.

• Grant greater autonomy to provincial governments, the recently created hospital unions, and hospitals themselves to manage secondary care services.

4. Using payments systems to encourage quality

• Further reforms in hospital payment mechanism to incentivise outcomes by introducing case-based payments in the hospital sector coupled with better monitoring of variation in clinical practice across reigns; special attention should go to ensuring that high volumes of hospital care do not endanger safety and effectiveness of care.

• Consider expanding the performance-based component of family physician remuneration to chronic care conditions such as diabetes, for example by introducing penalties for failure to monitor HbA1C levels in diabetic patients, and consider using payment to encourage activities such as care co-ordination, life style modification and other public health activities.

oecd reviews of health care quality: turkey · pdf fileoecd reviews of health care quality...

Documents